Perspectives in Computational Genome Analysis

DNA segments which together cover a genome may be collected together to form a genomic dictionary of specific words, which may be annotated either by biological information (according to the functional role they may play in regulatory mechanisms), or by numerical information (such as the position in the genome, the total number of occurrences, the occurrences lying inside or outside genic sequences, the CpG content, and more sophisticated informational indexes of text analysis). In this chapter, two analogous and complementary dictionary-based approaches to genome analysis are reviewed. We give a sketch of some of the relevant knowledge about the (human) genome, in terms of structure and functional role of its parts, and an informational view based on a mathematical analysis of k-mer dictionaries, with the aim of opening the way to the formulation of a model. Basic notions about genomic regulatory activity, where the underlying mechanisms of information exchange are far from understood, are given. A description of an initial attempt at computational modeling of genomes, seen as a new language to be deciphered, concludes the chapter.